{"id":"W4387828398","doi":"10.2196/51873","title":"Usability and Efficacy of Artificial Intelligence Chatbots (ChatGPT) for Health Sciences Students: Protocol for a Crossover Randomized Controlled Trial","year":2023,"lang":"en","type":"article","venue":"JMIR Research Protocols","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":47,"is_retracted":false,"has_abstract":true,"ca_institutions":"Dalhousie University; Carleton University; Centre Intégré Universitaire de Santé et de Services Sociaux du Centre-Sud-de-l'Île-de-Montréal; Université de Montréal; Centre for Interdisciplinary Research in Rehabilitation","funders":"","keywords":"Randomized controlled trial; Usability; Protocol (science); Chatbot; Computer science; Medical education; Crossover study; Medicine; Psychology; Alternative medicine; Artificial intelligence; Human–computer interaction","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.02862243,0.0002073315,0.001469032,0.0004947377,0.0007504047,0.0001943766,0.0003485135,0.0001634069,0.00006610725],"category_scores_gemma":[0.01630229,0.0001449268,0.0003646082,0.001011127,0.001555109,0.0001652821,0.0001246004,0.000313037,0.00001565344],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001661873,"about_ca_system_score_gemma":0.00204289,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001344759,"about_ca_topic_score_gemma":0.00003494463,"domain_scores_codex":[0.9941018,0.001023929,0.001936976,0.0006884168,0.001318268,0.0009305997],"domain_scores_gemma":[0.9882632,0.009259518,0.0004622223,0.000406073,0.001295575,0.0003134079],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"randomized_trial","study_design_gemma":"randomized_trial","study_design_scores_codex":[0.9487997,0.001908957,0.0002337728,0.002490552,0.0000432513,4.272996e-7,0.001716068,0.000007009024,0.0001435563,0.002860526,0.001152322,0.04064391],"study_design_scores_gemma":[0.8938959,0.02647581,0.0002271202,0.003900234,0.00001781538,0.000002444219,0.001856503,0.007642216,0.00975938,0.04728329,0.008646526,0.0002927425],"study_design_candidate":"randomized_trial","study_design_consensus":"randomized_trial","genre_codex":"protocol","genre_gemma":"protocol","genre_scores_codex":[0.01980797,0.000002691235,0.001190496,0.0042162,0.00004260991,0.9746169,0.00001431505,0.00006792103,0.00004084399],"genre_scores_gemma":[0.008090373,0.000001905799,0.0007763932,0.0001006837,0.000564209,0.9901614,0.000008611988,0.00002498151,0.0002715075],"genre_candidate":"protocol","genre_consensus":"protocol","teacher_disagreement_score":0.05490373,"threshold_uncertainty_score":0.9920018,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.6149109269353942,"score_gpt":0.7088118341520205,"score_spread":0.09390090721662636,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}